A method of increasing temporal resolution the method a) provides a current video having an initial spatial resolution and an initial temporal resolution, the current video has a temporally reduced video of an original video; b) repeatedly reduces the spatial resolution of the current video and the original video to produce a lowest spatial resolution current video and original video, the lowest spatial resolution current video having the initial temporal resolution; c) increases the temporal resolution of the lowest spatial resolution current video; d) increases the temporal resolution of the next higher spatial resolution current video; and e) repeats step (d) up to the initial spatial resolution. Steps c) and d) further include calculating enhancement data and using the enhancement data to enhance the increased temporal resolution of the current video, the calculating including using the respective reduced spatial resolution original video.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method of increasing temporal resolution, comprising: a. providing a current video having an initial spatial resolution and an initial temporal resolution, said current video comprising a temporally reduced video of an original video; b. repeatedly reducing the spatial resolution of the current video and the original video to produce a lowest spatial resolution current video and original video, said lowest spatial resolution current video having said initial temporal resolution; c. increasing the temporal resolution of said lowest spatial resolution current video as follows: i. Oracle prediction of a higher temporal resolution of said lowest spatial current video using temporal and/or spatial information of said lowest spatial current video; ii. calculating enhancement data and using said enhancement data to enhance the increased temporal resolution of said current video said calculating comprising using the respective reduced spatial resolution original video; d. increasing the temporal resolution of the next higher spatial resolution current video as follows: i. Oracle prediction of a higher temporal resolution of said next higher spatial resolution current video using temporal and/or spatial information of said higher temporal resolution current video and next higher spatial resolution current video and ii. calculating enhancement data and using said enhancement data to enhance the increased temporal resolution of said current video said calculating comprising using the respective reduced spatial resolution original video; and e. repeating step (d) up to the initial spatial resolution.
2. A method of video compression, comprising the method of claim 1 , wherein said current video comprises an already decoded temporally reduced video of the original video; wherein said operations on the original video are performed in an encoder; said Oracle prediction operations are performed in the decoder and simulated by the encoder; said calculating enhancement data is performed in the encoder; said enhancement data is provided from said encoder to a decoder; and said using the enhancement data is performed in both the encoder and the decoder.
3. The method of claim 2 , wherein said calculating enhancement data comprises analyzing the temporal motion field and the spatial geometrical structure of said current and original videos.
4. The method of claim 2 , further comprising, before reducing in step (b), blurring said current and original videos.
5. The method of claim 2 , wherein said Oracle prediction operations comprises analyzing the temporal motion field and the spatial geometrical structure of said current video.
6. The method of claim 1 , further comprising, before reducing in step (b), blurring said current and original videos.
7. The method of claim 1 , wherein said Oracle prediction operations in steps (c) and (d) comprises analyzing the temporal motion field and the spatial geometrical structure of said current video.
8. The method of claim 1 , wherein said calculating enhancement data comprises analyzing the temporal motion field and the spatial geometrical structure of said current and original videos.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
October 22, 2018
August 4, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.